Letter

What is punctuated equilibrium? What is

macroevolution? A response to Pennell et al.

1 2

Bruce S. Lieberman and Niles Eldredge

1

Department of Ecology and Evolutionary Biology and Biodiversity Institute, University of Kansas, Lawrence, KS, USA

2

Division of Paleontology, American Museum of Natural History, New York, NY, USA

What is punctuated equilibrium [or ‘equilibria’ as termed by higher-level selection (species selection) in shaping patterns

one of us (N.E.)]? What is macroevolution? These are ques- of diversity?’ Compare these with what PE actually states:

tions central to evolutionary biology. Yet, recently in an species are stable (show stasis) throughout much of their

article in Trends Ecol. Evol. by Pennell et al. [1] the former is evolutionary history; new species evolve cladogenetically

mischaracterized and the latter is not even defined. Here, we when a daughter population become geographically isolated

rectify these oversights by providing a definition of macro- from its parent species (via allopatric speciation); PE entails

evolution, an accurate rendition of punctuated equilibrium implicitly that both the ancestral species and daughter

(PE), and an elucidation of how they must be related. species coexist for at least some time after the speciation

Furthermore, consider the title of [1]: ‘Is there room for event [2–5]. These are the patterns predicted by PE (and

punctuated equilibrium. . .?’ Such a question is flawed: con- given the definition of macroevolution provided above, they

cepts do not battle for lebensraum, although sometimes their suggest a crucial link between PE and macroevolution).

opponents and proponents seem compelled to. Certainly, as part of PE, allopatric speciation is invoked.

However, many different processes can act when a popula-

Macroevolution defined tion becomes geographically isolated, running the gamut

We define macroevolution as: ‘the patterns and processes from selection to drift to neutral evolution, and so on; thus,

pertaining to the birth, death, and persistence of species’. many different processes can produce the pattern of clado-

Simpsonian definitions of macroevolution considered it genesis in allopatry required by PE. Furthermore, let us

evolution above the species level. However, with such a consider the pattern of stasis. Again, many different pro-

definition, either macroevolution becomes meaningless, cesses can act to produce stasis in a lineage [4–6].

because there is no evolutionary process operating above Given these statements about the actual, original, and

the species level (i.e., no ‘generification’) or macroevolution subsequently generally used and intended meanings of PE,

is simply the study of pattern. By failing to define macro- let us first return to three of the four points ascribed to PE

evolution, Pennell et al. fail to frame the scope of the in [1]. Points (i) and (ii) are statements about pattern and

problem they are considering. are not two distinct statements but instead intimately

related ones. Gradualistic evolution necessitates anage-

What PE really is netic change. Pulsed evolution implies cladogenetic change

We focus on three additional flaws in [1]. Two involve failure except when an entire species transforms en masse into

to define PE accurately and distinguish between the pattern another species. Even if speciation ever happens this way,

of PE and the process(es) that produce it. The third is that it is not a model of speciation consistent with allopatric

Pennell et al. claim implications of the theory that are not speciation invoked by PE. Therefore, point (i) presented in

correct. They admit they are not trying to find ‘the true [1] is a pattern that can be ascribed to PE. However, point

‘essence’ of PE’ ([1] p. 24), but this limits the validity of their (ii) is not entirely accurate because it either is synonymous

conclusions, by allowing them to define PE any way they with (i) or conflates the mechanism of speciation invoked by

wish. They do so in a manner inconsistent with usage by the PE. How does Pennell et al.’s point (iii) fare? PE states that

original framers of PE, and by numerous other subsequent speciation occurs allopatrically, but many different pro-

authors. Pennell et al. ([1] p. 24) ascribe four research cesses can transpire in allopatry to cause population diver-

questions to PE: ‘(i) what is the relative importance of gence. Therefore, point (iii) of [1] is a statement about

gradualistic versus pulsed evolution? (ii) what is the role processes of allopatric speciation, demonstrating the

of speciational events (cladogenesis) versus within lineage notion that Pennell et al. conflate pattern with process

evolution (anagenesis) in generating trait divergence? (iii) in their consideration of PE. To the extent that Mayr [7],

when change is cladogenetic, are the changes adaptive or the source for Eldredge and Gould’s mechanism of specia-

driven by neutral processes? and (iv) how important is tion [2], invoked nonadaptive factors, we can suppose that

PE included the possibility of these, but it is never expli-

Corresponding author: Lieberman, B.S. ([email protected]).

citly stated as a requirement of PE (e.g., [2–5]). Certainly,

0169-5347/$ – see front matter

Gould and others [5,8] argued that adaptation need not be

ß 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.tree.2014.02.005

invoked to explain every episode of diversification. How-

ever, most of these discussions were not in the context of

PE, but were part of discussions in the adaptive radiations

literature. Nowhere was it suggested in [2] or in any paper

Trends in Ecology & Evolution, April 2014, Vol. 29, No. 4 185

Letter Trends in Ecology & Evolution April 2014, Vol. 29, No. 4

by Eldredge or Gould authored together or separately on ‘species sorting’, which can sometimes involve the process

PE (or in many other subsequent papers that considered of species selection. In any case, differential survival and

PE; e.g., [6,9]) that, for PE to be validated, speciation generation of species remains a pattern entailed by PE, but

events must involve nonadaptive factors. Thus, point species selection is not entailed. Therefore, Pennell et al.’s

(iii) of [1] conflates the pattern and the process of PE point (iv) does not deal directly with PE; failure to docu-

and is an erroneous extrapolation of its implications. ment species selection does not refute PE.

In summation, it is not PE that conflates four separate

Species selection is not required by PE primary research questions, but Pennell et al.’s interpreta-

Finally, we turn to Pennell et al.’s point (iv), which focused tion and extrapolation of PE that does. Still, we agree that

on the importance of species selection. The authors argued a ‘truly synthetic macroevolutionary research program will

that, for PE to hold, species selection must be an important involve the melding of data and theory from different

evolutionary force. However, as they acknowledge, species disciplines’ ([1] p. 30) and is desirable.

selection has long been a controversial topic. In fact, let us

consider their definition ([1] p. 29]) of species selection.

Acknowledgments

They acknowledge that it ‘ignore(s)’ the distinction

We thank Mark Pagel for helpful comments.

between species selection and sorting made by Vrba and

Gould! [10]. Indeed, the definition of species selection that References

Pennell et al. use does not match the definition of species 1 Pennell, M.W. et al. (2013) Is there room for punctuated equilibrium in

selection used in many publications, including those macroevolution? Trends Ecol. Evol. 29, 23–32

2 Eldredge, N. and Gould, S.J. (1972) Punctuated equilibria: an

authored or coauthored by the architects of PE [4,10,11].

alternative to phyletic gradualism. In Models in Paleobiology

Indeed, various authors have argued that species selection

(Schopf, T., ed.), pp. 82–115, Freeman Cooper

is unlikely to have a major role in evolution [4,10,11], yet

3 Gould, S.J. and Eldredge, N. (1977) Punctuated equilibria: the tempo

these authors endorsed PE. It is true that, in some pub- and mode of evolution reconsidered. Paleobiology 3, 115–151

lications (e.g., [5]), Gould used a definition of species selec- 4 Eldredge, N. (1989) Macroevolutionary Dynamics, McGraw Hill

5 Gould, S.J. (2002) The Structure of Evolutionary Theory, Harvard

tion corresponding to that used by Pennell et al. [1], but he

University Press

waffled on this issue [5,11]. Furthermore, he never stated

6 Eldredge, N. et al. (2005) The dynamics of evolutionary stasis.

that, for PE to be validated, species selection must occur.

Paleobiology 31, 133–145

(Gould [5] argued that documenting species selection was 7 Mayr, E. (1963) Animal Species and Evolution, Harvard University

important for the independence of macroevolutionary the- Press

8 Gould, S.J. and Vrba, E.S. (1982) Exaptation: a missing term in the

ory.) In addition, it is clear from papers authored or

science of form. Paleobiology 8, 4–15

coauthored by the architects of PE that species selection

9 Lieberman, B.S. et al. (2007) Paleontological patterns, macroecological

never had to be a prominent evolutionary force for PE to dynamics and the evolutionary process. Evol. Biol. 34, 28–48

prevail. Indeed, the only source that we could find for this 10 Vrba, E.S. and Gould, S.J. (1986) The hierarchical expansion of sorting

and selection: sorting and selection cannot be equated. Paleobiology 12,

was [1]. By contrast, Eldredge and Gould [2] stated that,

217–228

with PE, explaining trends required differential genera-

11 Lieberman, B.S. and Vrba, E.S. (2005) Stephen Jay Gould on species

tion and survival of species; Stanley [12] referred to this

selection: 30 years of insight. Paleobiology 31, 113–131

pattern as species selection. Contra [12], various authors 12 Stanley, S.M. (1975) A theory of evolution above the species level. Proc.

[4,9–11] have argued that this pattern should be called Natl. Acad. Sci. U.S.A. 72, 646–650

186